Available online at www.sciencedirect.

com

ScienceDirect
Underground Space xxx (xxxx) xxx
www.elsevier.com/locate/undsp

Analysis of a tunnel failure caused by leakage of the shield tail

seal system
Chao Yu a, Annan Zhou b,⇑, Jun Chen a, Arul Arulrajah c, Suksun Horpibulsuk d
a
State Key Laboratory of Ocean Engineering, Department of Civil Engineering, School of Naval Architecture, Ocean and Civil Engineering, Shanghai
Jiao Tong University, Shanghai 200240, China
b
Discipline of Civil and Infrastructure Engineering, School of Engineering, Royal Melbourne Institute of Technology (RMIT), Victoria 3001, Australia
c
Department of Civil and Construction Engineering, Swinburne University of Technology, Victoria 3122, Australia
d
School of Civil Engineering, Suranaree University of Technology, Nakhon Ratchasima 30000, Thailand

Received 10 September 2018; received in revised form 16 November 2018; accepted 19 November 2018

Abstract

This paper presents an analysis of a tunnel failure accident during shield tunnel construction on Foshan Metro Line 2 in China. The
failure is caused by the leakage of the multilayer seal system, which consists of several brush seals at the tail of the shield. Four different
failure modes for the multilayer seal system are discussed. A simple structural analysis of the brush seals is then conducted, and failure
mode 4 (failure due to brush seal deformation) is identified as a major reason for the Foshan tunnel accident. A Finite Element Method
(FEM) analysis is employed to validate the conclusions drawn from the simple structural analysis of the brush seals.
Ó 2018 Tongji University and Tongji University Press. Production and hosting by Elsevier B.V. on behalf of Owner. This is an open access article
under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Keywords: Leakage; Shield tunneling; Brush seals; Finite element

1 Introduction places (Shen, Cui, Ho, & Xu, 2016; Zhao, Peng, Wang,
Zhang, & Jiang, 2016). When shield tunneling is employed
Shield tunneling is widely used for tunnel construction in complex geological strata with rich ground water (Wu
in urban areas because of fewer disturbances to the sur- et al., 2016, 2019; Shen, Wu, & Misra, 2017; Wu, Shen,
roundings compared with traditional mining tunneling & Yang, 2017; Wu, Shen, Cheng, & Hino, 2017; Wu,
(Liao, Liu, Wang, & Li, 2009; Shen et al., 2009, 2010; Shen, Yang, & Zhou, 2018; Lyu, Xu, Cheng, &
Wu, Shen, & Liao, 2015; Ren, Shen, & Arulrajah et al., Arulrajah (2018)), such as mixed strata with soft sticky soil
2018; Chai, Shen, & Yuan, 2018; Lyu, Wu, Shen, & (Jin, Yin, Wu, & Daouadji, 2018; Jin, Yin, Wu, & Zhou,
Zhou (2018)). Shield tunneling can also be employed in 2018; Yin, Jin, Shen, & Hicher, 2018; Yin, Wu, &
various geological conditions, such as soft deposits, weak Hicher, 2018) and hard rock (Ren, Shen, & Chai et al.,
rocks and mixed grounds (Zhang, Shen, Zhou, & 2018; Yin, Hicher, & Dano, 2017), groundwater may rush
Arulrajah, 2018; Ren, Shen, & Chai et al., 2018; Ren, into the shield from either the cutter face or the tail (Liu,
Shen, Arulrajah, & Cheng (2018)). During shield tunneling, Shen, Xu, & Yin, 2018; Lyu, Wang, Shen, Lu, & Wang,
the cutter face and the tail are usually identified as weak 2016, Lyu, Wang, Cheng, & Shen, 2017; Xu, Shen, Lai,
& Zhou (2018) and Xu, Shen, Ren, & Wu (2016)). Water
leakage in the shield may cause failure of the tunnel and
affect the surrounding environment (Wongsaroj, Soga, &
⇑ Corresponding author.
Mair, 2007, 2013; Xu, Shen, Ma, Sun, & Yin, 2014;
E-mail address: annan.zhou@rmit.edu.au (A. Zhou).

https://doi.org/10.1016/j.undsp.2018.11.003
2467-9674/Ó 2018 Tongji University and Tongji University Press. Production and hosting by Elsevier B.V. on behalf of Owner.
This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
2 C. Yu et al. / Underground Space xxx (xxxx) xxx

Soga, Laver, & Li, 2017; Lyu, Sun, Shen, & Arulrajah, 2 Analysis of the seal brush performance in different
2018, Lyu, Shen, & Arulrajah, 2018; Tan & Lu, 2018; conditions
Ren, Xu, Shen, & Zhou (2018)). This paper reports a tun-
nel failure that occurred at the shield tail, which was caused 2.1 The multilayer seal system
by groundwater leakage in Foshan Metro Line 2, China
(Investigation Committee, 2018). Figure 3 shows a picture of the tail of the shield
Foshan Metro Line 2 (Project I) is an east-west metro machine. As shown in Fig. 3, three circles of brush seals
line that includes 12 underground stations and 5 elevated are installed and welded between the shield machine and
stations. The project commenced in August 2014, and segments around the inner wall of the shield machine
the total planned investment is 18.94 billion Chinese (Shi, Zhao, Yang, Ju, & Yu, 2015; Liu et al., 2018; Lyu,
Yuan (approximately $2.99 billion US dollars). The Shen, Zhou, & Yang, 2019). Figure 4 shows a schematic
length of Foshan Metro Line 2 is 32.4 km, and it starts diagram of a shield tail and a picture of a brush seal.
in Nan Zhuang and ends at the Guangzhou South Rail- The gaps between different brush seals are filled with grease
way Station in Foshan City (see Fig. 1), Guangdong (Figure 4). Through the grease, three circles of brush seals
Province, China. The project has five shield tunnel inter- form a multilayer seal system that resists the external
vals, and some of the intervals were constructed in strata groundwater pressure, the synchronous grouting pressure,
that feature rich underground water and complex geolog- and the pressure caused by the settlement of the founda-
ical conditions. The tunnel interval between Huchong tion. The grease should completely fill the gaps to ensure
Station and Lvdaohu Station (i.e., the H-L interval tun- effective pressure transmission between the brush seal rings.
nel) in the project has a length of 1 804 m and a depth The parameters of the main components of the brush
of 17.2~34.4 m (Investigation Committee, 2018). seals are shown in Table 1. The conditions of the surround-
On the evening of February 7, 2018, a ground col- ing soils of the accident area are shown in Fig. 5. The thick-
lapse incident (urban sink hole) occurred during the con- nesses and parameters of the surrounding soil layers are
struction of the H-L interval tunnel. The size of the sink listed in Table 2.
hole was approximately 900 m2, and the depth was
approximately 6 m (see Fig. 2); the accident caused 11 2.2 Failure modes of the multilayer seal system
deaths and 8 injuries and resulted in 1 missing person.
The accident was attributed to a leakage at the tail of Leakage accidents occur at the shield machine tail if the
the shield machine, which was not sealed well by the multilayer seal system fails. If the surroundings are mainly
brush seals. composed of soil with a strong fluidity, such as fine sands,
Brush seals are usually employed to resist the external leakage accidents will also cause a more severe progressive
pressure, which consists of the groundwater pressure, the failure. Such a progressive failure may eventually lead to
synchronous grouting pressure, and the pressure caused the crash of the shield and even a ground collapse (similar
by the settlement of the foundation. The performance of to the accident in Foshan Metro Line 2). According to the
the brush seals under the external pressure conditions is structural features and mechanical analysis of the multi-
analyzed in this paper, and four different failure modes of layer seal system in some engineering accident instances
brush seals are summarized based on the analysis. Two (Zheng et al., 2017; Han, Ye, Chen, Xia, & Wang, 2017),
safety evaluation indexes are proposed to interpret the the four different failure modes of the system under differ-
mechanism of multilayer brush seals. ent loading types are summarized in Table 3.

Fig. 1. (a) Location of Foshan City in China; (b) Location of Foshan Metro Line 2.

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
 C. Yu et al. / Underground Space xxx (xxxx) xxx 3

Fig. 2. Top view of the sink hole.

Fig. 3. The picture of the tail of the shield machine and brush seals.

Fig. 4. The structural diagram of a shield tail and a picture of a brush seal.

3 Structural analysis of the brush seals to uniform distributed loads that reflect its geometri-
cal shape and force conditions. The critical external
As shown in Fig. 6, since the circle of brush seals pressure load and critical displacement can be esti-
is round and symmetrical, any one of the brush seals mated from the simplified brush seal model as
can be simplified as an indeterminate beam subjected follows.

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
4 C. Yu et al. / Underground Space xxx (xxxx) xxx

Table 1
Parameters of the main components of the brush seals.
Main Components Size Weight (kg)
Front stainless-steel wire brushes £0.35 mm, 170 mm
200 mm 0.40
£0.35 mm, 200 mm
200 mm 0.48
Rear stainless-steel wire brushes £0.35 mm, 220 mm
200 mm 0.62
£0.35 mm, 250 mm
200 mm 0.70
Front protection plate 160 mm
200 mm
0.6 mm 0.18
Rear protection plate 240 mm
200 mm
1 mm 0.54
Shell plating 125 mm
200 mm
5 mm 1.36
90 mm
200 mm
5 mm

Fig. 5. The distribution of surrounding soil layers in the accident area of Foshan Metro Line 2.

Table 2
Thicknesses and parameters of the surrounding soil layers.
Soil Layer H (m) q (kN/m3) C (kPa) ui (°) E0 (MPa) m
1. Backfill 3.2 17.05 18 10 3.52
2. Silty soil 18 16.17 9.11 17.61 2.45 0.37
3. Soft plastic silty clay 3.4 18.82 23 16 5.59 0.31
4. Fine sand 7.4 18.23 25 7.5 0.3
5. Round gravel 2.8 20.09 35 18 0.22
6. Highly weathered argillaceous sandstone 9 22.54 80 28 80 0.21
Note: H is the thickness of each soil layer, q is the density, C is the effective cohesion, ui is the effective frictional angle, E0 is the modulus of deformation,
and m is the Poisson’s ratio.

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
 C. Yu et al. / Underground Space xxx (xxxx) xxx 5

Table 3
Characteristics of the different failure modes for the multilayer seal systems.
Failure modes Characteristics Consequence
Mode 1: puncture failure  High external pressure punctures some of the May lead to a progressive failure at the tail of the
brush seals shield machine
 The integrity of the whole seal system is
broken
Mode 2: interlayer failure  Low grease pressure weakens the sealing Only the outside circle of brush seals is still
between brush seals capable of sealing.
 The multilayer seal system is at risk
Mode 3: failure due to ground deformation  The ground heave caused by over grouting Local leakage happens at the gap
produces a gap between segments at the tail
of the shield machine
Mode 4: failure due to brush seal deformation  Damage of the brush seals results in the failure May produce a progressive failure at the tail of the
of the multilayer seal system. shield machine
 Damage of the brush seals may be attributed
to the (i) incorrect positioning of the shield
machine, (ii) dislocation of the shield seg-
ments, and (iii) abrasion of the steel wire
brushes

Fig. 6. The geometry diagram of a brush seal and its force conditions.

3.1 Simplified calculation model of the brush seals where b is the width of the brush seal (m) and h is its thick-
ness (m).
As shown in Fig. 6, the simplified brush seal is subjected Since the bending moment at point A is the maximum
to an external pressure from the left side and the hydraulic and the section modulus in bending at point A is the min-
pressure from the grease on the right side. The simplified imum, the bending normal stress of the simplified brush
brush seal is roller-supported at point A and fixed at point seal at point A reaches its maximum, which is
B with a tilt angle. The maximum bending moment (M A )
can be achieved at point A, and its magnitude is MA
r¼ ð3Þ
Wz
pl2
MA ¼ cos a ð1Þ
8 From Eqs. (1)–(3), the resultant pressure p can be
expressed as
where p is the resultant pressure of the external pressure
and hydraulic pressure of the grease (MPa), l is the length 4rbh2
p¼ ð4Þ
of the simplified brush seal (m), and a is the inclined angle 3l2 cos a
between the simplified brush seal and the perpendicular line
(°). The multilayer seal system at the tail of the shield
The section modulus (W z ) in bending at point A is machine consists of three circles of brush seals with grease
in the gaps. For simplicity, the pressure carried by each cir-
1 cle is assumed to be identical, and the resultant pressure is
W Z ¼ bh2 ð2Þ
6 assumed to be the sum of the pressures on each single-layer

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
6 C. Yu et al. / Underground Space xxx (xxxx) xxx

brush seal. Therefore, the designed external pressure pd is To quantify the designed external pressure, it is neces-
equal to three times the resultant pressure p calculated sary to take the geometrical and mechanical parameters
from each brush seal, namely of the simplified brush seal into account, which are listed
in Table 4.
According to Eqs. (4) and (5) and Table 3, the designed
pd ¼ 3p ð5Þ external pressure is calculated as pd = 1.190 MPa.

Table 4
Parameters of the simplified brush seal.
Parameters Size E (MPa) rs (MPa) a (°)
Magnitude 250 mm
200 mm
18 mm 342.7 235 35
Note: E is the strength of extension, rs is the yield strength, and a is the included angle.

Fig. 7. Mesh of the numerical model in Plaxis.

Fig. 8. Overall deformation of tunnels with complete and damaged seal systems: (a) tunnel deformation with a complete seal system; (b) tunnel
deformation with a damaged seal system.

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
 C. Yu et al. / Underground Space xxx (xxxx) xxx 7

Fig. 9. Distribution of the axial forces (N), shearing forces (Q), and bending moments (M) for (a) Scenario I—a complete multilayer seal system—and (b)
Scenario II, which is a damaged multilayer seal system.

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
8 C. Yu et al. / Underground Space xxx (xxxx) xxx

Table 5 4 Numerical evaluation

Additional parameters used for the numerical model.
0
D (m) hs (m) H (m) E (GPa) m pw (MPa) ps (MPa) 4.1 Numerical model
6.7 0.2 25 210 0.3 0.22 0.4
Note: D is the tunnel diameter, hs is the brush seal thickness, H is the
The numerical model (see Figs. 7 and 8) of the tail of the
0
buried depth of the tunnel, E is the elastic modulus of the brush seals, m is shield is built in Plaxis to analyze the influence of the con-
the Poisson’s ratio of the brush seals, pw is the groundwater pressure, and dition of the seal system on the stress and moment distribu-
ps isthesynchronous grouting pressure. tion in the tunnel with the same external pressure state. The
following two scenarios are considered here. In the first sce-
nario (Scenario I), the multilayer seal system at the tail of
3.2 Safety evaluation indexes the shield machine is assumed to be complete. In the sec-
ond scenario (Scenario II), the system is assumed to be
According to the failure modes listed in Table 3, failure damaged (with a leakage point at 210° on the circle, see
modes 1 and 4 of the multilayer seal system are more harm- Fig. 9) under the same external pressure. The brush seal
ful to the safety of the tunnel because they have more sev- element at the leakage point features a high permeability
ere consequences. Therefore, the first and second and a low shear strength. The surrounding soil condition
evaluation indexes are proposed to evaluate the safety of (see Table 2) in the accident area of Foshan Metro Line
the system against failure modes 1 and 4, respectively. 2 is used in these two scenarios. Table 4 shows the addi-
The first evaluation index for the safety of the multilayer tional parameters and their values, which are used in the
seal system is the pressure ratio (np ), which is defined as numerical model.
follows.
p 4.2 Failure mode
np ¼ r ð6Þ
pd
where pr is the external pressure outside the brush seals According to the parameters from the numerical simula-
(MPa) and pd is the designed external pressure (MPa). tion in Table 5, the pressure ratio (np ) of both scenarios is
The external pressure pr is usually equal to the sum of equal to 0.52, which is safe because the value of np is less
the groundwater pressure, the synchronous grouting pres- than one. The FEM results indicate that failure mode 1 is
sure and the other external pressures. The seal is safe not expected to occur for Foshan Metro Line 2. This con-
against failure mode 1 if the value of the pressure ratio clusion is identical to that based on the simple structural
(np ) is less than one. analysis presented in Section 3.2.
The second evaluation index for the safety of the multi- The overall deformations for the two scenarios are pre-
layer seal system is the ovality (Dh ), which is defined as sented in Fig. 8. As shown in Fig. 8, the ovality (Dh ) for
follows: Scenario I is 0.36%, which meets the security requirement
  (0.5%); however, the ovality (Dh ) of Scenario II is larger
Dmax  Dmin than 5%, which far exceeds 0.5%. These results clearly indi-
Dh ¼ 2 ð7Þ
Dmax þ Dmin cate that Scenario I is safe in terms of the second evalua-
where Dmax and Dmin are the maximum and the minimum tion index but Scenario II is not. Therefore, the FEM
outer diameters of the circle of the brush seals (m), respec- results indicate that failure mode 4 is the probable failure
tively. The seal is regarded to be safe against failure mode 4 mode for the accident that happened in Foshan Metro Line
if the value of the ovality (Dh ) is less than 0.5%. 2. This conclusion is identical to that based on the simple
In the accident area of Foshan Metro Line 2, the exter- structural analysis presented in Section 3.2.
nal groundwater pressure is approximately 0.3 MPa, and
the average value of the synchronous grouting pressure is 4.3 Force and moment distributions
0.4 MPa. The real external pressure is approximately
0.7 MPa, and the pressure ratio, np ¼ ppr ¼ 0:58 < 1. There- Dysfunction and/or breakage of one or more brush seals
d
fore, failure model 1 is not expected to happen at the tail of in a circular seal system lead to a leakage point. The leak-
the shield for this case. For Foshan Metro Line 2, the outer age causes a stress redistribution around the shield machine
diameter of the brush seals is 6.7 m. Assuming the sum of tail. The analysis shows that the axial forces, shearing
the maximum and the minimum outer diameters, forces and bending moments all feature a concentration
Dmax þ Dmin , is equal to 2 times the normal outer diameter, phenomenon due to the leakage (see Fig. 9). At one partic-
the sum should be 13.4 m. The measured diameter differ- ular point, the bending moment in Scenario II is 3 times
ence ðDmax  Dmin Þ around the accident area is approxi- higher than that in Scenario I. Such a high stress concen-
mately 39 mm. Therefore, Dh is approximately 0.58%, tration may lead to further damage/breakage of the seg-
which is slightly larger than 0.5%. Therefore, failure mode ments and other structures in the tunnel shield. If the
4 is suspected to be the possible failure mode of the acci- surrounding soil contains a large amount of small grains
dent of Foshan Metro Line 2 in this case. or shows high fluidity, a progressive failure could happen

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
 C. Yu et al. / Underground Space xxx (xxxx) xxx 9

that causes the subsequent failure of the whole tunnel, Lyu, H. M., Wang, G. F., Shen, J. S., Lu, L. H., & Wang, G. Q. (2016).
Analysis and GIS mapping of flooding hazards on 10 May, 2016,
which causes a ground collapse incident. Guangzhou, China. Water, 8(10), 447.
Lyu, H. M., Shen, S. L., Zhou, A. N., & Yang, J. (2019). Perspectives for
5 Conclusions flood risk assessment and management for mega-city metro system.
Tunneling and Underground Space Technology, 84, 31–44.
Ren, D. J., Shen, S. L., Arulrajah, A., & Wu, H. N. (2018). Evaluation of
The tail failure during shield tunneling in Foshan Metro ground loss ratio with moving trajectories induced in double-O-tube
Line 2, China, is investigated in this paper. Through a sim- (DOT) tunnelling. Canadian Geotechnical Journal, 55(6), 894–902.
Ren, D. J., Shen, J. S., Chai, J. C., & Zhou, A. N. (2018). Analysis of disk
ple structural analysis of the brush seals, the failure mode cutter failure in shield tunnelling using 3D circular cutting theory.
for the Foshan tunnel accident has been identified as a fail- Engineering Failure Analysis, 90, 23–35.
ure due to the deformation of the brush seals. The FEM Ren, D. J., Shen, S. L., Arulrajah, A., & Cheng, W. C. (2018). Prediction
model of TBM disc cutter failures during tunnelling in mixed-face
analysis then validated the conclusion drawn from the sim- ground. Rock Mechanics and Rock Engineering, 51(11), 3599–3611.
ple structural analysis of the brush seals. The analyses also Ren, D. J., Xu, Y. S., Shen, J. S., & Zhou, A. N. (2018). Prediction of
indicate that the tunnel shield can be damaged due to the ground deformation during pipe-jacking considering multiple factors.
Applied Sciences, 8(7), 1051.
leakage at the tails because it causes a substantial concen- Shen, S. L., Cui, Q. L., Ho, C. E., & Xu, Y. S. (2016). Ground response to
tration of the bending moments, axial forces and shearing multiple parallel microtunneling operations in cemented silty clay and
forces at the leakage area. sand. Journal of Geotechnical and Geoenvironmental Engineering, 142
(5), 04016001.
Shen, S. L., Du, Y. J., & Luo, C. Y. (2010). Evaluation of the effect of
Acknowledgments rolling-correction of double-o-tunnel shields via one-side loading.
Canadian Geotechnical Journal, 47(10), 1060–1070.
Shen, S. L., Horpibulsuk, S., Liao, S. M., & Peng, F. L. (2009). Analysis of
The research work described herein was funded by the the behavior of DOT tunnel lining caused by rolling correction
National Basic Research Program of China (973 Program: operation. Tunneling and Underground Space Technology, 24(1), 84–95.
2015CB057806). This financial support is gratefully Shen, S. L., Wu, Y. X., & Misra, A. (2017). Calculation of head difference
at two sides of a cut-off barrier during excavation dewatering.
acknowledged. Computers and Geotechnics, 91, 192–202.
Shi, J., Zhao, Y., Yang, Q., Ju, Y., & Yu, C. (2015). Application of
grouting and water-blocking techniques to the replacement of shield
References tail seal brushes in a permeable strata with high water pressure.
Modern Tunnelling Technology, 52(4), 190–194.
Chai, J. C., Shen, J. S., & Yuan, D. J. (2018). Mechanism of tunneling- Soga, K., Laver, R. G., & Li, Z. L. (2017). Long-term tunnel behavior and
induced cave-in of a busy road in Fukuoka city, Japan. Underground ground movements after tunneling in clayey soils. Underground Space,
Space, 3(2), 140–149. 2(3), 149–167.
Han, L., Ye, G. L., Chen, J. J., Xia, X. H., & Wang, J. H. (2017). Tan, Y., & Lu, Y. (2018). Responses of shallowly buried pipelines to
Pressures on the lining of a large shield tunnel with a small adjacent deep excavations in Shanghai soft ground. Journal of Pipeline
overburden: A case study. Tunnelling & Underground Space Tech- Systems Engineering and Practice, 9(2), 05018002.
nology, 64, 1–9. Wongsaroj, J., Soga, K., & Mair, R. J. (2013). Tunnelling-induced
Investigation Committee (2018). Investigation report on ‘‘2.7” water consolidation settlements in London Clay. Geotechnique, 63(13),
leakage and ground collapse major accident in Project I of Foshan 1103–1115.
Metro Line 2 in Guangdong, the people’s Government of Guangdong Wongsaroj, J., Soga, K., & Mair, R. J. (2007). Modelling of long-term
Province of China (in Chinese) ground response to tunnelling under St James’s Park, London.
Jin, Y. F., Yin, Z. Y., Wu, Z. X., & Daouadji, A. (2018). Numerical Geotechnique, 57(1), 75–90.
modeling of pile penetration in silica sands considering the effect of Wu, H. N., Shen, S. L., & Liao, S. M. (2015). Longitudinal structural
grain breakage. Finite Elements in Analysis and Design, 144, 15–29. modelling of shield tunnels considering shearing dislocation between
Jin, Y. F., Yin, Z. Y., Wu, Z. X., & Zhou, W. H. (2018). Identifying segmental rings. Tunnelling and Underground Space Technology, 50,
parameters of easily crushable sand and application to offshore pile 317–323.
driving. Ocean Engineering, 154, 416–429. Wu, H. N., Shen, S. L., & Yang, J. (2017a). Identification of tunnel
Liao, S. M., Liu, J. H., Wang, R. L., & Li, Z. M. (2009). Shield tunneling settlement caused by land subsidence in soft deposit of Shanghai.
and environment protection in Shanghai soft ground. Tunneling and Journal of Performance of Constructed Facilities, ASCE, 31(6),
Underground Space Technology, 24(4), 454–465. 04017092.
Liu, X. X., Shen, S. L., Xu, Y. S., & Yin, Z. Y. (2018). Analytical Wu, Y. X., Shen, J. S., Cheng, W. C., & Hino, T. (2017). Semi-analytical
approach for time-dependent groundwater inflow into shield tunnel solution to pumping test data with barrier, wellbore storage, and
face in confined aquifer. International Journal for Numerical and partial penetration effects. Engineering Geology, 226, 44–51.
Analytical Methods in Geomechanics, 42, 655–673. Wu, H. N., Shen, S. L., Yang, J., & Zhou, A. N. (2018). Soil-tunnel
Lyu, H. M., Sun, W. J., Shen, S. L., & Arulrajah, A. (2018). Flood risk interaction modelling for shield tunnels considering shearing disloca-
assessment in metro systems of mega-cities using a GIS-based tion in longitudinal joints. Tunneling and Underground Space Technol-
modeling approach. Science of the Total Environment, 626, 1012–1025. ogy, 78, 168–177.
Lyu, H. M., Shen, S. L., & Arulrajah, A. (2018). Assessment of Wu, Y. X., Shen, S. L., & Yuan, D. J. (2016). Characteristics of
geohazards and preventive countermeasures using AHP incorporated dewatering induced drawdown curve under blocking effect of retaining
with GIS in Lanzhou, China. Sustainability, 10(2), 304. wall in aquifer. Journal of Hydrology, 539, 554–566.
Lyu, H. M., Xu, Y. S., Cheng, W. C., & Arulrajah, A. (2018). Flooding Wu, Y. X., Lyu, H. M., Han, J., & Shen, S. L. (2019). Case study:
hazards across southern China and prospective sustainability mea- Dewatering-induced building settlement around a deep excavation in
sures. Sustainability, 10(5), 1682. the soft deposit of Tianjin, China. Journal of Geotechnical and
Lyu, H. M., Wu, Y. X., Shen, J. S., & Zhou, A. N. (2018). Assessment of Geoenvironmental Engineering, ASCE (accepted for publication).
social-economic risk of Chinese dual land use system using fuzzy AHP. Xu, Y. S., Shen, S. L., Lai, Y., & Zhou, A. N. (2018). Design of sponge
Sustainability, 10(7), 2451. city: Lessons learnt from an ancient drainage system in Ganzhou,
Lyu, H. M., Wang, G. F., Cheng, W. C., & Shen, S. L. (2017). Tornado China. Journal of Hydrology, 563, 900–908.
hazards on June 23rd in Jiangsu Province, China: Preliminary Xu, Y. S., Shen, S. L., Ma, L., Sun, W. J., & Yin, Z. Y. (2014). Evaluation
investigation and analysis. Natural Hazards, 85(1), 597–604. of the blocking effect of retaining walls on groundwater seepage in

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003
10 C. Yu et al. / Underground Space xxx (xxxx) xxx

aquifers with different insertion depths. Engineering Geology, 183, tive function. Journal of Engineering Mechanics, ASCE, 144(4),
254–264. 04018014.
Xu, Y. S., Shen, S. L., Ren, D. J., & Wu, H. N. (2016). Analysis of factors Zhang, N., Shen, J. S., Zhou, A. N., & Arulrajah, A. (2018). Tunneling
in land subsidence in Shanghai: A view based on Strategic Environ- induced geohazards in mylonite rock fault with rich groundwater: A
mental Assessment. Sustainability, 8(6), 573. case study in Guangzhou. Tunneling and Underground Space Technol-
Yin, Z. Y., Hicher, P. Y., Dano, C., & Jin, Y. F. (2017). Modeling the ogy, 74, 262–272.
mechanical behavior of very coarse granular materials. Journal of Zhao, J. W., Peng, F. L., Wang, T. Q., Zhang, X. Y., & Jiang, B. N.
Engineering Mechanics, 143(1), C4016006. (2016). Advances in master planning of urban underground space
Yin, Z. Y., Jin, Y. F., Shen, J. S., & Hicher, P. Y. (2018). Optimization (UUS) in China. Tunnelling and Underground Space Technology, 55,
techniques for identifying soil parameters in geotechnical engineering: 390–1307.
Comparative study and enhancement. International Journal for Zheng, G., Cui, T., Cheng, X., Diao, Y., Zhang, T., Sun, J., et al.
Numerical and Analytical Methods in Geomechanics, 42(2), 70–94. (2017). Study of the collapse mechanism of shield tunnels due to the
Yin, Z. Y., Wu, Z. Y., & Hicher, P. Y. (2018). Modeling the monotonic failure of segments in sandy ground. Engineering Failure Analysis,
and cyclic behavior of granular materials by an exponential constitu- 79, 464–490.

Please cite this article as: C. Yu, A. Zhou, J. Chen et al., Analysis of a tunnel failure caused by leakage of the shield tail seal system, Underground Space,
https://doi.org/10.1016/j.undsp.2018.11.003